Although previous work showed that visualizations could be used for communication in clinical settings, their main focus has been on using visualizations to find reoccurring patterns and detect outliers. Our work extends this literature to uncover features of visualizations that are useful for communicating about ASD through a series of studies on two behavior visualizations, Plexlines and EnGaze.
Plexlines uses color coded circles on a timeline to visualize three different modalities of child behaviors. Plexlines was designed to display communicative behaviors and to aid in identifying developmental delay in the domain of ASD. The diameter of the circle is proportional to the duration of the annotated behavior, which accentuates behaviors with longer duration, such as prolonged eye contact with the examiner. The visualization also offers additional features for its users, such as providing an aggregate view to see patterns over multiple visualizations and allowing users to customize the visualization by removing certain behaviors based on their specific needs. The webtool offers customization capabilities such as filtering, sorting, and hiding certain behaviors. A user study conducted with ASD researchers and clinicians revealed that Plexlines has potential to be integrated into existing behavioral evaluation processes, aid in the detection of developmental delays in young children, and serve as a visual artifact to better communicate with parents.
EnGaze visualizes communicative behaviors, but by using colored rectangles on a timeline. The unique characteristics of EnGaze is the inclusion of detailed examiner's behaviors and its focus on identifying moments of joint attention during an RABC session. Joint attention is a set of communicative behaviors that signals the shared focus of two individuals, and the lack of joint attention is a defining characteristic of ASD. Thus, identifying joint attention, or the lack thereof, can lead to early detection of ASD, and in turn contribute to early intervention. EnGaze provides several features that highlight moments of joint attention. Users can highlight moments when the child responds to the examiner, and customize the visualization through hiding user selected behaviors in the visualization.
H. Kong, J. Lee, and K. Karahalios. A Comparative Study of Visualizations with Different Granularities of Behavior for Communicating about Autism, Proceedings of the ACM on Human-Computer Interaction:CSCW 2018. pdf
J. Lee, H. Kong, S. Lin, and K. Karahalios, Plexlines: Tracking Socio-communicative Behaviors Using Timeline Visualizations, AMIA 2016. pdf
H. Kong, J. Lee, and K. Karahalios. EnGaze: Designing Behavior Visualizations with and for Behavioral Scientists, DIS 2016. pdf